metasyn.distribution.uniform

Uniform distributions and fitters.

Classes

`BaseDTUniformDistribution`(lower, upper[, ...])	Base class for all time/date/datetime uniform distributions.
`BaseDTUniformFitter`(privacy)	Base class for date/time/datetime uniform distributions.
`ContinuousUniformDistribution`(lower, upper)	Uniform distribution for floating point type.
`ContinuousUniformFitter`(privacy)	Fitter for continuous uniform distribution.
`DateTimeUniformDistribution`(lower, upper[, ...])	Uniform DateTime distribution.
`DateTimeUniformFitter`(privacy)	Fitter for datetime uniform distribution.
`DateUniformDistribution`(lower, upper)	Uniform date distribution.
`DateUniformFitter`(privacy)	Fitter for date uniform distribution.
`DiscreteUniformDistribution`(lower, upper)	Uniform discrete distribution.
`DiscreteUniformFitter`(privacy)	Fitter for discrete uniform distribution.
`TimeUniformDistribution`(lower, upper[, ...])	Uniform time distribution.
`TimeUniformFitter`(privacy)	Fitter for time uniform distribution.

class metasyn.distribution.uniform.BaseDTUniformDistribution(lower, upper, precision='microseconds')

Bases: BaseDistribution

Base class for all time/date/datetime uniform distributions.

Parameters:

lower (Any)
upper (Any)
precision (str)

round(time_obj)

Round down any time object with the precision.

Parameters:: time_obj (Any) – Object to round down.
Returns:: Time/date/datetime object rounded down to the measured precision.
Return type:: obj

draw()

Draw a random element from the fitted distribution.

Return type:: datetime

abstractmethod fromisoformat(dt_obj)

Convert string to iso format.

Parameters:: dt_obj (str)

property minimum_delta: timedelta: Get the minimum time delta.

information_criterion(values)

Get the BIC value for a particular set of values.

Parameters:: values (array_like) – Values to determine the BIC value of.

class metasyn.distribution.uniform.BaseDTUniformFitter(privacy)

Bases: BaseFitter

Base class for date/time/datetime uniform distributions.

Parameters:: privacy (BasePrivacy)

class metasyn.distribution.uniform.ContinuousUniformDistribution(lower, upper)

Bases: ScipyDistribution

Uniform distribution for floating point type.

This class implements the uniform distribution between a minimum and maximum.

Parameters:

lower (float) – Lower bound for uniform distribution.
upper (float) – Upper bound for uniform distribution.

Examples

>>> UniformDistribution(lower=-3.0, upper=10.0)

name: core.uniform

unique: False

version: 1.0

var_type: continuous

classmethod default_distribution(var_type=None)

Get a distribution with default parameters.

Return type:: BaseDistribution

name: str = 'core.uniform': The identifier for the implemented distribution

var_type: Union[str, Sequence[str]] = 'continuous': The variable type of the distribution

class metasyn.distribution.uniform.ContinuousUniformFitter(privacy)

Bases: BaseFitter

Fitter for continuous uniform distribution.

Parameters:: privacy (BasePrivacy)

dist_class: <class ‘metasyn.distribution.uniform.ContinuousUniformDistribution’>

version: 1.0

var_type: continuous

privacy: none

distribution: alias of ContinuousUniformDistribution

class metasyn.distribution.uniform.DateTimeUniformDistribution(lower, upper, precision='microseconds')

Bases: BaseDTUniformDistribution

Uniform DateTime distribution.

Datetime objects will be uniformly distributed between a start and end date time.

Parameters:

lower (str or datetime.datetime) – Earliest possible datetime to be generated from this distribution. String formatted in ISO format, see examples below or use datetime objects from the python standard library datetime.
upper (str or datetime.datetime) – Latest possible datetime to be generated from this distribution. String formatted in ISO format, see examples below or use datetime objects from the python standard library datetime.
precision (str) – Most precise measure of datetime that the distribution will take into account. For example, if precision == “seconds”, then the microseconds will always be set to 0. Possible values: [“microseconds”, “seconds”, “minutes”, “hours”, “days”].

Examples

>>> DateTimeUniformDistribution(lower="2022-07-15T10:39", upper="2022-08-15T10:39",
                                precision="minutes")

name: core.uniform

unique: False

version: 1.0

var_type: datetime

fromisoformat(dt_obj)

Convert string to iso format.

Return type:: datetime
Parameters:: dt_obj (str)

classmethod default_distribution(var_type=None)

Get a distribution with default parameters.

Return type:: BaseDistribution

name: str = 'core.uniform': The identifier for the implemented distribution

var_type: Union[str, Sequence[str]] = 'datetime': The variable type of the distribution

class metasyn.distribution.uniform.DateTimeUniformFitter(privacy)

Bases: BaseDTUniformFitter

Fitter for datetime uniform distribution.

Parameters:: privacy (BasePrivacy)

dist_class: <class ‘metasyn.distribution.uniform.DateTimeUniformDistribution’>

version: 1.0

var_type: datetime

privacy: none

distribution: alias of DateTimeUniformDistribution

class metasyn.distribution.uniform.DateUniformDistribution(lower, upper)

Bases: BaseDTUniformDistribution

Uniform date distribution.

Date objects will be uniformly distributed between a start and end date.

Parameters:

lower (str or datetime.date) – Earliest possible date to be generated from this distribution. String formatted in ISO format, see examples below or use date objects from the python standard library datetime.
upper (str or datetime.date) – Latest possible date to be generated from this distribution. String formatted in ISO format, see examples below or use date objects from the python standard library datetime.

Examples

>>> DateUniformDistribution(lower="10:39:12", upper="10:39:45", precision="seconds")

name: core.uniform

unique: False

version: 1.0

var_type: date

fromisoformat(dt_obj)

Convert string to iso format.

Return type:: date
Parameters:: dt_obj (str)

round(time_obj)

Round down any time object with the precision.

Parameters:: time_obj – Object to round down.
Returns:: Time/date/datetime object rounded down to the measured precision.
Return type:: obj

classmethod default_distribution(var_type=None)

Get a distribution with default parameters.

Return type:: BaseDistribution

name: str = 'core.uniform': The identifier for the implemented distribution

var_type: Union[str, Sequence[str]] = 'date': The variable type of the distribution

class metasyn.distribution.uniform.DateUniformFitter(privacy)

Bases: BaseDTUniformFitter

Fitter for date uniform distribution.

Parameters:: privacy (BasePrivacy)

dist_class: <class ‘metasyn.distribution.uniform.DateUniformDistribution’>

version: 1.0

var_type: date

privacy: none

distribution: alias of DateUniformDistribution

class metasyn.distribution.uniform.DiscreteUniformDistribution(lower, upper)

Bases: ScipyDistribution

Uniform discrete distribution.

It differs from the floating point uniform distribution by being a discrete distribution instead.

Parameters:

lower (int) – Lower bound (inclusive) of the uniform distribution.
upper (int) – Upper bound (exclusive) of the uniform distribution.

Examples

>>> DiscreteUniformDistribution(lower=3, upper=20)

name: core.uniform

unique: False

version: 1.0

var_type: discrete

classmethod default_distribution(var_type=None)

Get a distribution with default parameters.

Return type:: BaseDistribution

name: str = 'core.uniform': The identifier for the implemented distribution

var_type: Union[str, Sequence[str]] = 'discrete': The variable type of the distribution

class metasyn.distribution.uniform.DiscreteUniformFitter(privacy)

Bases: BaseFitter

Fitter for discrete uniform distribution.

Parameters:: privacy (BasePrivacy)

dist_class: <class ‘metasyn.distribution.uniform.DiscreteUniformDistribution’>

version: 1.0

var_type: discrete

privacy: none

distribution: alias of DiscreteUniformDistribution

class metasyn.distribution.uniform.TimeUniformDistribution(lower, upper, precision='microseconds')

Bases: BaseDTUniformDistribution

Uniform time distribution.

Time objects will be uniformly distributed between a start and end time.

Parameters:

lower (str or datetime.time) – Earliest possible time to be generated from this distribution. String formatted in ISO format, see examples below or use time objects from the python standard library datetime.
upper (str or datetime.time) – Latest possible time to be generated from this distribution. String formatted in ISO format, see examples below or use time objects from the python standard library datetime.
precision (str) – Most precise measure of time that the distribution will take into account. For example, if precision == “seconds”, then the microseconds will always be set to 0. Possible values: [“microseconds”, “seconds”, “minutes”, “hours”].

Examples

>>> TimeUniformDistribution(lower="10:39:12", upper="10:39:45", precision="seconds")

name: core.uniform

unique: False

version: 1.0

var_type: time

fromisoformat(dt_obj)

Convert string to iso format.

Return type:: time
Parameters:: dt_obj (str)

classmethod default_distribution(var_type=None)

Get a distribution with default parameters.

Return type:: BaseDistribution

draw(): Draw a random element from the fitted distribution.

name: str = 'core.uniform': The identifier for the implemented distribution

var_type: Union[str, Sequence[str]] = 'time': The variable type of the distribution

class metasyn.distribution.uniform.TimeUniformFitter(privacy)

Bases: BaseDTUniformFitter

Fitter for time uniform distribution.

Parameters:: privacy (BasePrivacy)

dist_class: <class ‘metasyn.distribution.uniform.TimeUniformDistribution’>

version: 1.0

var_type: time

privacy: none

distribution: alias of TimeUniformDistribution